Efficient Dialogue State Tracking by Selectively Overwriting Memory

Abstract

Recent works in dialogue state tracking (DST) focus on an open vocabulary-based setting to resolve scalability and generalization issues of the predefined ontology-based approaches. However, they are inefficient in that they predict the dialogue state at every turn from scratch. Here, we consider dialogue state as an explicit fixed-sized memory and propose a selectively overwriting mechanism for more efficient DST. This mechanism consists of two steps: (1) predicting state operation on each of the memory slots, and (2) overwriting the memory with new values, of which only a few are generated according to the predicted state operations. Our method decomposes DST into two sub-tasks and guides the decoder to focus only on one of the tasks, thus reducing the burden of the decoder. This enhances the effectiveness of training and DST performance. Our SOM-DST (Selectively Overwriting Memory for Dialogue State Tracking) model achieves state-of-the-art joint goal accuracy with 51.72% in MultiWOZ 2.0 and 53.01% in MultiWOZ 2.1 in an open vocabulary-based DST setting. In addition, we analyze the accuracy gaps between the current and the ground truth-given situations and suggest that it is a promising direction to improve state operation prediction to boost the DST performance.

Accepted to ACL2020 as a long paper

References in corpus (5)

• A Network-based End-to-End Trainable Task-oriented Dialogue System
• NSML: Meet the MLaaS platform with a real-world case study
• Global-Locally Self-Attentive Dialogue State Tracker
• NSML: A Machine Learning Platform That Enables You to Focus on Your Models
• Non-Autoregressive Dialog State Tracking

Cited by in corpus (9)

• Efficient Context and Schema Fusion Networks for Multi-Domain Dialogue State Tracking
• MinTL: Minimalist Transfer Learning for Task-Oriented Dialogue Systems
• CREDIT: Coarse-to-Fine Sequence Generation for Dialogue State Tracking
• A Fast and Robust BERT-based Dialogue State Tracker for Schema-Guided Dialogue Dataset
• A Contextual Hierarchical Attention Network with Adaptive Objective for Dialogue State Tracking
• Leveraging Slot Descriptions for Zero-Shot Cross-Domain Dialogue State Tracking
• Zero-Shot Dialogue State Tracking via Cross-Task Transfer
• Modeling Long Context for Task-Oriented Dialogue State Generation
• Improving Limited Labeled Dialogue State Tracking with Self-Supervision